ifixai-ai/iFixAi
Runs 32 automated checks on your AI agent and grades it A-F across five pillars.
iFixAi audits an AI agent's real-world behavior rather than just its technical stats — it runs 32 inspections across five pillars and returns an A-F grade with a scored scorecard in under two minutes. You can drive it as a guided CLI wizard, with explicit CLI flags for CI, or as a plugin/skill inside Claude Code, Cursor, or Codex, where the agent configures and explains the run itself. API keys are auto-detected and stored by environment-variable name, never as the secret itself.
What holds up
- +Full run finishes in under two minutes, per the project — fast enough to check before shipping a change.
- +Three ways to run it — guided wizard, scripted CLI flags, or an in-agent plugin — fit different workflows.
- +Stores only the environment-variable name for API keys in its config, never the key itself.
Mind the limits
- −Needs your own provider API key (OpenAI, Anthropic, Gemini, etc.) — grading runs on an LLM you pay for.
- −CLI and plugin based, not a hosted dashboard — you install via pip or a CLI plugin to use it.
Featured here? Take the badge
Put it on your site — it links back to this review. Free for every listed product, always.
<a href="https://stillworks.dev/products/p/ifixai-ai-ifixai/"><img src="https://stillworks.dev/badge/ifixai-ai-ifixai.svg" alt="Picked by StillWorks" width="250" height="54"></a>